Paper Info
Title
An efficient approach for self-scheduling parallel loops on multiprogrammed parallel computers
Abstract
Clusters and grids have increasingly become standard platforms for high performance computing as they provide extremely high execution rates with great cost effectiveness. Such systems are designed to support concurrent execution of multiple jobs. It calls for multiprogrammed scheduling of the different jobs for effective system utilization and for keeping average response times low. Although a significant amount of work has been done in scheduling parallel jobs on multiprocessor systems, the problem of scheduling parallel tasks of an individual job on a multiprogrammed parallel system has not been given enough attention so far. In this paper, we present a dynamic scheduling technique for scheduling iterations of a DOALL loop (of a single application) to achieve load balance between a given set of processors. Experimental results show the effectiveness of our approach.
Year
DOI
Venue
2005
10.1007/978-3-540-69330-7_31
LCPC
Keywords
Field
DocType
efficient approach,multiprogrammed scheduling,multiprogrammed parallel computer,great cost effectiveness,parallel task,high execution rate,parallel job,dynamic scheduling technique,multiprogrammed parallel system,high performance computing,parallel loop,concurrent execution,effective system utilization,load balance,dynamic scheduling,parallel computer,parallel systems
Multiprocessor scheduling,Fair-share scheduling,Supercomputer,Computer science,Scheduling (computing),Parallel computing,Multiprocessing,Two-level scheduling,Rate-monotonic scheduling,Dynamic priority scheduling,Distributed computing
Conference
Volume
ISSN
ISBN
4339
0302-9743
3-540-69329-7
Citations 
PageRank 
References 
2
0.39
6
Authors
3
Name
Order
Citations
PageRank
Arun Kejariwal128126.23
Alexandru Nicolau22265307.74
Constantine D. Polychronopoulos3893129.02